Custom Layers

There are two components to adding a custom layer:
1.
Adding the layer configuration class: extends org.deeplearning4j.nn.conf.layers.Layer
2.
Adding the layer implementation class: implements org.deeplearning4j.nn.api.Layer
The configuration layer ((1) above) class handles the settings. It's the one you would use when constructing a MultiLayerNetwork or ComputationGraph. You can add custom settings here, and use them in your layer.
The implementation layer ((2) above) class has parameters, and handles network forward pass, backpropagation, etc. It is created from the org.deeplearning4j.nn.conf.layers.Layer.instantiate(...) method. In other words: the instantiate method is how we go from the configuration to the implementation; MultiLayerNetwork or ComputationGraph will call this method when initializing the
An example of these are CustomLayer (the configuration class) and CustomLayerImpl (the implementation class). Both of these classes have extensive comments regarding their methods.
You'll note that in Deeplearning4j there are two DenseLayer clases, two GravesLSTM classes, etc: the reason is because one is for the configuration, one is for the implementation. We have not followed this "same name" pattern here to hopefully avoid confusion.
Testing Your Custom Layer
Once you have added a custom layer, it is necessary to run some tests to ensure it is correct.
These tests should at a minimum include the following:
1.
Tests to ensure that the JSON configuration (to/from JSON) works correctly
This is necessary for networks with your custom layer to function with both
model serialization (saving) and Spark training.
2.
Gradient checks to ensure that the implementation is correct.
Example
A full custom layer example is available in our examples repository.

Deeplearning4j - Previous

How To Guides

Keras Import

Export as PDF

Copy link

Contents

Testing Your Custom Layer

Example